I have written on this blog, over the years, a fair bit about models. This has included their functions, purposes, limitations and uses, along with more obscure subjects such as the ethics of their use and the ability of a single model to represent reality, at least in a fairly limited manner. In all this, however, I have left the question of why we need models, or why we have models, rather to one side.
Now, of course, models pop up in all sorts of places in life. For example, science and engineering could hardly progress without models, at least of a certain kind. To some extent, a scientific model is a paradigm of what a model should be. It is specific, it give predictive answers for given input. Its range of applicability can even be defined, so that it is known when a specific model might not give reliable answers. And, of course, the model can be tested against the real world, to see if the assumptions and approximations made in constructing it are valid.
Other forms of model also appear in life, however. For example, in social sciences there is a secularization model, or approach, which makes certain assumptions about the impact of scientific and technological advances on society, and the disenchantment thereof. This is not, of course, a mathematical model, but it allows a certain broad based interpretation of evidence to be placed within a particular schema. Nor is the secularisation model predictive, except in the widest sense that what has happened in the past is likely to continue into the future.
The advantage of such models is that they can, at least, be argued over. The secularisation model is not, now, widely accepted to apply (as religion has made a comeback in culture, society and politics) but it still gives an interpretation of evidence, even though that interpretation is not now widely accepted.
In history, too, models are used, although they are not, perhaps, widely acknowledged as such. For example, some of the arguments about the origins of the English Civil War rely on models, such as the rise of the gentry, the decline of the gentry, the impossibility of governing three disparate kingdoms, the problems of having a king like Charles I, and so on. Each of these is, or implies, a certain model or set of models of the natures of the societies and cultures of the time, and the fact that they do not agree, or contradict each other, implies that here are interesting questions to tackle.
A model, then, in the social sciences or arts and humanities is not, strictly speaking, then, a scientific model, but an interpretative tool of… well, what?
The point about models is that they allow us to focus on certain bits of a system. By assumption, some bits are believed to be important to the topic in hand, and some bits are not. Thus, in writing say, a set of wargame rules for a given period, we might argue that the interaction of pikemen and musketeers is vital, and focus our modelling effort on that aspect. We would then evolve a set of tactical rules which, if we had done our job well, would model the musketeers occasionally running away and hiding under the lowered pikes of the central unit block.
Now, of course, this model could well be challenged. Evidence for musketeers actually behaving in this manner might be deemed to be weak, and other models, such as disciplined musketeers discharging a salvo at fifty paces and seeing off the horse, could be put forward. But this is, more or less, the function of the model in the first place. The original model provokes, through questioning of its function and assumptions, a counter model to be developed.
This is all a bit Hegelian, of course. We have a thesis, followed by an antithesis, and we look forwards to some sort of synthesis, at which point the whole cycle starts again, with our new synthesis model provoking something different. But the point is that this progress only appears as a result of working with the particular models involved.
There is, then, something about the model and its use which provokes the human mind into action. We study the models, even if they are, in the case of history, implied ones. We examine their assumptions and try to question them, bearing in mind that the assumptions made may either be very close to our own if we come from the same culture and society, or very different. We can then try to scan the evidence put forward for the correct functioning of the model. How many gentry families were in decline in the decades before 1640? How many were prospering? And so on.
By thus working with the model, its outcomes and evidences, we can start to draw our own conclusions. The model provides us with mental stimulation, with questions to ask about understanding of the model and what it is trying to represent. If you have ever tried to explain something to someone, eventually you have to leave it up to the person in question. Eventually we run out of explanations, and the person has to work it out for themselves. Hopefully, our explanations will have been sufficiently good for them to exclaim ‘Ah, now I’ve got it’. This is the moment of insight, the point at which the model discloses something about the world which the mind can grasp.
The point, therefore, about models is that they give the mind a simplified universe to examine, in which disclosures about what is (or was) ‘really’ going on occur. If a wargame is true to some sort of reality, then it should be able to provoke some sort of disclosure to the participants. I think this would work even for a non-historical scenario. Medieval French knights could well be swamped by Incas in a wargame, which might disclose to the wargamers that medieval French knights needed proper support, and were not a super-weapon in and of themselves.
And, of course, if that happens, we might have learned something to feed back into our rules.